Bug 623468 - Incorrect hwtable entry for HSVX700
Summary: Incorrect hwtable entry for HSVX700
Alias: None
Product: Red Hat Enterprise Linux 4
Classification: Red Hat
Component: device-mapper-multipath
Version: 4.8
Hardware: x86_64
OS: Linux
Target Milestone: rc
: ---
Assignee: Ben Marzinski
QA Contact: Gris Ge
Depends On:
TreeView+ depends on / blocked
Reported: 2010-08-11 20:35 UTC by Sebastian Koehler
Modified: 2015-09-16 23:46 UTC (History)
25 users (show)

Fixed In Version: device-mapper-multipath-0.4.5-42.el4
Doc Type: Bug Fix
Doc Text:
Clone Of: 567760
Last Closed: 2011-02-16 14:26:13 UTC
Target Upstream Version:

Attachments (Terms of Use)
multipath -v3 (9.06 KB, text/plain)
2010-08-11 20:35 UTC, Sebastian Koehler
no flags Details
/etc/multipath.conf (2.20 KB, text/plain)
2010-08-11 20:36 UTC, Sebastian Koehler
no flags Details

System ID Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2011:0243 normal SHIPPED_LIVE device-mapper-multipath bug fix and enhancement update 2011-02-15 16:34:54 UTC

Description Sebastian Koehler 2010-08-11 20:35:25 UTC
Created attachment 438295 [details]
multipath -v3

+++ This bug was initially created as a clone of Bug #567760 +++

I was able to reproduce the issue on RHEL4 U8. 

[root@linux ~]# cat /etc/*release
Red Hat Enterprise Linux ES release 4 (Nahant Update 8)
[root@linux ~]# uname -a
Linux linux.local 2.6.9-89.0.28.ELsmp #1 SMP Thu Jul 22 18:14:41 EDT 2010 x86_64 x86_64 x86_64 GNU/Linux

I've attached the multipath.conf and the multipath -v3 output for your reference. With my modifications to multipath.conf the LUN failover is working fine. I'm just curious about the "faulty" status of the passive paths.

[root@linux ~]# multipath -ll
mpath2 (3600a0b80001111550000da0bc987624c)
[size=25 GB][features="0"][hwhandler="0"]
\_ round-robin 0 [prio=8][active]
 \_ 0:0:6:1 sdi 8:128 [active][ready]
 \_ 1:0:7:1 sdl 8:176 [active][ready]
\_ round-robin 0 [enabled]
 \_ 0:0:7:1 sdj 8:144 [active][faulty]
 \_ 1:0:6:1 sdk 8:160 [active][faulty]

Comment 1 Sebastian Koehler 2010-08-11 20:36:37 UTC
Created attachment 438296 [details]

Comment 2 Sebastian Koehler 2010-08-11 20:37:54 UTC
The SVSP release used is 3.02 which is still active/passive.

Comment 4 Ben Marzinski 2010-09-21 14:25:57 UTC
The config that seems to work best is

device {
               vendor                  "HP"
               product                 "HSVX700"
               path_grouping_policy    group_by_prio
               prio_callout            "/sbin/mpath_prio_hp_sw /dev/%n"
               hardware_handler        "0"
               path_checker            hp_sw
               failback                immediate
               no_path_retry           12
               rr_min_io               100

Comment 5 Ben Marzinski 2010-10-28 03:29:29 UTC
autoconfiguration added.

Comment 7 Gris Ge 2011-01-18 10:19:34 UTC

/usr/share/doc/device-mapper-multipath-0.4.5/multipath.conf.defaults is not updated.

multipathd -k'show config' cannot work in RHEL4, is there any way for me to check the autoconfiguration?

Thank you.

Comment 8 Gris Ge 2011-01-18 10:20:04 UTC

In device-mapper-multipath-0.4.5-41.el4:
/usr/share/doc/device-mapper-multipath-0.4.5/multipath.conf.defaults is not updated.

multipathd -k'show config' cannot work in RHEL4, is there any way for me to check the autoconfiguration?

Thank you.

Comment 13 Ben Marzinski 2011-01-25 16:10:42 UTC
The old config has been removed, and the documentation has been updated.

Comment 16 Gris Ge 2011-01-26 07:17:58 UTC
Both multipath.conf.defaults and libmultipath/hwtable.c have been updated in device-mapper-multipath-0.4.5-42.el4

Code reviewed. Sanity Only.

Comment 17 errata-xmlrpc 2011-02-16 14:26:13 UTC
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.


Note You need to log in before you can comment on or make changes to this bug.